UK turns to science to test new policies

2012-07-04 20:34

London - Would you pay a fine more promptly if you received a personalised text message reminder? Would you settle your accounts sooner if you got a letter from the taxman saying most people in your area had already paid up?

As austerity hits government spending in Britain, academics are helping policymakers answer questions like these, applying scientific method to improve efficiency and squeeze every last drop out of departments' limited budgets.

Randomised controlled trials - where scientists give one groups a placebo and another group the new intervention being tested, then watch for any difference - have been used for decades in medicine and are now starting to appear in the corridors of power.

"In medicine it is considered to be almost perverse not to do such a test. You'd be required to subject a medicine to a trial like this before introducing it to a market," said Owain Service, Deputy Director of the Behavioural Insights Team at the Cabinet Office, which helps the British government put policies into action.

"In areas like international development it's increasingly prevalent. Foreign governments say: we want to know if the money that we're spending is having the effect that we desire of it. The best way of understanding that is to run a randomised controlled trial.

"Essentially what we're saying is that there's no reason why large areas of public policy in the developed world can't also use this approach."

The Behavioural Insights Team, led by experimental psychologist Dr Laura Haynes, was set up two years ago to build on growing evidence from behavioural economics and psychology that subtle changes in policies can have big impacts.

Reminders pay

In a report published in June, the Behavioural Insights Team recommended more policymakers take a leaf out of scientists' books and use randomised controlled trials to determine if a policy works.

It cited one study that had already looked into the impact of text messages on people who owed court fines.

The trial found that sending a reminder text significantly increased the likelihood of prompt repayment, while personalising the text could bring in an estimated further £3m a year at minimal cost if rolled out nationally.

The team is now working with job centres to look at testing back-to-work schemes - signing up one group of workers to a new programme and comparing their employment prospects with another group that carried on with the existing one.

The concept has its critics, too, the report acknowledged. Some politicians told researchers the tests were too expensive - and that they worried about the moral implications of depriving one group of people of a potentially helpful policy, just to set up a control group.

The report argued the benefits of testing outweighed the possible costs, and it was better to deprive a small number initially than spend money on a scheme that was ineffective or even harmful.

"I'd be lying if I said every government department we have discussions with embraces it," said Service.

"But there's definitely been a sense in the last twelve months or so that this is something that we can and should be doing in far greater areas of government policy."

This includes "almost every domestic department", Service said, including the Department of Business which is looking at conducting trials on access to finance for small businesses. In the limp British economy, where credit remains tight, business budgets are also hurting.

"It's more important than ever to understand what value for money different interventions are providing," Service said.